Table 1 Direct human interventions dominate the non-stationarity responses in peakflow across different datasets and metrics in the US for the long term (> 60 years)

From: Direct human interventions drive spatial variability in long-term peak streamflow trends across the United States

Non-stationarity metrics

All USGS sites [3907]

Climate Change (reference only)

Direct human interventions +

Climate change

(non-reference)

HCDN

[439]

GAGESII- reference [620]

GAGES II non-reference [2830]

Other USGS

[457]

Trend

Increasing

503 (12.9%)

65 (14.8%)

85 (13.7%)

363 (12.8%)

55 (12.0%)

Decreasing

820 (21.0%)

27 (6.2%)

39 (6.3%)

638 (22.5%)

143 (31.3%)

Field Significant Trend

Increasing

302 (7.7%)

28 (6.4%)

36 (5.8%)

232 (8.2%)

34 (7.4%)

Decreasing

624 (16%)

12 (2.7%)

16 (2.6%)

490 (17.3%)

118 (25.8%)

Variance

Unequal Variance

674 (17.3%)

34 (7.7%)

46 (7.4%)

525 (18.6%)

103 (22.5%)

Stationarity

Non-stationary

1195 (30.6%)

65 (14.8%)

91 (14.7%)

917 (32.4%)

187 (40.9%)

  1. This table displays the number of USGS sites demonstrating statistically significant changes (p-value less than 0.05) in different datasets (columns) and the non-stationarity metrics (rows). The statistical methods employed included the Mann-Kendall test for trend, the False Discovery Rate test for the field significance test of trends, the Levene test for variance, and the Kwiatkowski-Phillips-Schmidt-Shin (KPSS) test for stationarity (see Method). The numbers in the square bracket and the header row represent the total count of USGS sites within the respective dataset. The numbers in parentheses indicate the percentage of total sites that exhibited significant changes. Notably, there was remarkable consistency in the analysis outcomes across diverse datasets and metrics.
Back to article page